Artificial occultation?

[Thalestris24]

Just read this interesting piece by Phil Plait. He's using the Hubble ST but wondered whether anyone knew of a way amateurs could occult a bright star in order to image fainter stuff that might be close by? It doesn't sound an easy thing to do but thought I'd ask just in case!

Louise

[vlaiv]

I think that blocking the star can be achieved relatively easy, but I'm not sure it will have wanted benefit of being able to image nearby things (at least not faint ones).

It would go something like this:

Use eyepiece projection instead of prime focus. Use some sort of clear glass filter placed in focal plane of eyepiece with a single blocking dot - size that of a star image for particular scope - you can think of reticle eye piece but instead of having crosshair - having blocker that is sized so it covers star image (star airy disk - couple of microns across).

Problem with this setup is that it will block the star airy disk - it will not block out diffracted light that forms glare around bright stars - for that you really need to block light before it enters objective of the scope and becomes diffracted. From what I've gathered from the article, they actually imaged similar star in order to remove diffraction halo (subtracted halo signal) - while this will help in removing the halo signal it will not help with remaining noise (Poisson / shot noise from all that light in halo), hence it will not be suitable for imaging really faint stuff close to star (like exoplanets) because SNR will be really poor.

[vlaiv]

I just thought of another method to achieve probably even better results, without using occultation at all.

We can split effects of bright star into two parts:

1. Core - this is the part that needs occulting in case one is using CCD that does not have antiblooming gate, since all those electrons gathered from the bright star can start spilling over to adjacent pixels.

2. Halo - this is larger section around core that is usually obstructing faint stuff that we want to image.

Now there is no real boundary between 1 and 2, and what we call core can increase in diameter due to seeing and guiding errors. If one uses camera with ABG and appropriate sub duration, electron spillage is not an issue. Halo still remains an issue, so while we really do not need to block the core, we need to do something about the halo. I think there is a way to deal with halo without blocking the light. Halo extends in circular fashion around the core because scope aperture is circular. By using different aperture configuration we can influence where halo light ends up. If we use square aperture we can force the halo light into spikes (like spider spikes) - so the whole diffracted light from aperture ends up in spikes and not in circular halo. If we rotate mask in such way that spikes don't cover object that we are imaging, there is good chance that good SNR on faint object can be achieved. If we don't know exact position of faint object - we can take set exposures with one orientation of spikes, and other set of exposures with different orientation of spikes, and choose the set that does not have spikes covering the object.

I planned to experiment with this approach by shooting close doubles (trying to see if I can go beyond the theoretical resolution limit of scope / seeing by using lucky imaging, custom stacking algorithm and square aperture mask)

[Thalestris24]

Hi

Thanks for the input! The specialised Hubble camera uses a rotating bar - obviously not something an amateur can easily replicate. A faint object nearby will need a relatively long exposure which would normally result in star bloat. I was thinking of maybe some sort of reticule next to/on the sensor so that it had a dot(s) the same size the star would be without bloat. On the other hand, it's probably much too much trouble for something that's not likely to crop up very often. It'll probably have to remain a thought experiment!

Louise

[SonnyE]

Sounds more like some sort of Artificial Eclipsation...